Page 3 of 7 FirstFirst 12345 ... LastLast
Results 31 to 45 of 93

Thread: What's wrong with this picture??

  1. #31
    Join Date
    Aug 2012
    Location
    Pacific Palisades, California
    Posts
    73
    Quote Originally Posted by Mike Duffey View Post
    Jeff, the title of that thread was, "Johnny Miller in his prime, "second fire" in the downswing. Is this a movement you would teach? How would you guess this shows up in a kinematic sequence graph?".
    If Kelvin never meant for the second fire to related to pelvis angular velocity - which I'm sure he knows is what the kinematic sequence shows - then why did he title it that way? If he knew it was something different, then there is no purpose in asking if/how it would show up in the KS.

    I'm sure you are familiar with his Skype session with Tapio. You only need to work through the first five minutes to hear them going through the kinematic sequence. Tapio brings up the apparent second acceleration of the shoulders - which is very clearly and specifically a rotational velocity - and Kelvin refers to that as a "Second Fire". He makes no effort to clarify that he is talking about a non-rotational motion. A minute or so earlier, when the topic is pelvis rotation in the KS, he makes no effort to mention that his second fire would be a different kind of motion.

    In fact, in that thread when Kelvin was asked why he added some lines to the video, he responded,
    "What are the yellow lines for? To help those that cannot see 2D movement with their own eyes. It's up to the biomechanists to explain this movement to us. We just make observations. Where in the current graphs is this motion buried? Why do our observations not jibe with the deceleration shown in the graphs?"

    I have to think that Kelvin understands the different between rotation and linear translation. If not, he is an idiot who has no business teaching anything in golf. I believe he is not an idiot. So he keeps asserting that this motion should be seen in a graph of rotational velocity.

    So, maybe he knew and maybe he didn't. Reading through the thread, it become eminently clear that the people in the conversation originally understood the Kelvin's second fire to be a rotational motion, not a linear one, and that understanding changed over time. Kelvin made no effort early on to clear up the topic.

    What is absolutely clear is that many of the early comments were made under the belief that Kelvin was discussing a second rotational velocity peak. My comment about not taking them out of context is that the discussion here, as it did in the original thread, has shifted to a linear motion. I agree that you have not reposted them, I'm trying to save anyone who starts reading this discussion now the headache that we went through as the conversation shifted from a rotational to a linear motion discussion.

    As for putting together a good compilation of reading material, why not do it? It seems it would only help people understand the topic.
    Dear Mike, Jeff and others passionately interested in this VERY beneficial exchange,

    Jeff, great to see you posting so much TPI/AMM data and analyses after so much apparent dependence on just high speed video. I hope it indicates that you see/understand the benefits of these other sensing systems.

    Mike, I can see why I have not been able to talk to you from the depth and content of these exchanges. Clearly my stuff regarding dynamic balance and control can wait.

    Both of you, I would like to suggest a pseudo scientific approach to the 'second fire' understanding that works for me. The idea/analogy comes from the launch monitor world where there has been a 'posting' of both 'back spin AND side spin. I hope there can be TOTAL agreement that the golf ball, after launch/impact with the club spins around only one axis until impact with the ground .Well, in my world of dynamic balance and control, I have gotten comfortable looking for and establishing for each body element motion, an 'instantaneous and apparent' screw axis of rotation.

    Not as an olive branch, but as a sincere suggestion for the pelvis/hip rotation discussions, can you, as I have been able to, envision 3D 'curved' paths of points in the pelvic area, that result from both the aforementioned rotary and linear motions ??. And isn't this like the golf ball's single axis of rotation what is really happening ??

  2. #32
    Mike Duffey Guest
    Quote Originally Posted by art View Post
    Dear Mike, Jeff and others passionately interested in this VERY beneficial exchange,

    Jeff, great to see you posting so much TPI/AMM data and analyses after so much apparent dependence on just high speed video. I hope it indicates that you see/understand the benefits of these other sensing systems.

    Mike, I can see why I have not been able to talk to you from the depth and content of these exchanges. Clearly my stuff regarding dynamic balance and control can wait.

    Both of you, I would like to suggest a pseudo scientific approach to the 'second fire' understanding that works for me. The idea/analogy comes from the launch monitor world where there has been a 'posting' of both 'back spin AND side spin. I hope there can be TOTAL agreement that the golf ball, after launch/impact with the club spins around only one axis until impact with the ground .Well, in my world of dynamic balance and control, I have gotten comfortable looking for and establishing for each body element motion, an 'instantaneous and apparent' screw axis of rotation.

    Not as an olive branch, but as a sincere suggestion for the pelvis/hip rotation discussions, can you, as I have been able to, envision 3D 'curved' paths of points in the pelvic area, that result from both the aforementioned rotary and linear motions ??. And isn't this like the golf ball's single axis of rotation what is really happening ??
    Art,
    Let's use the ball example for a moment (no pun intended). There are two reasons why I prefer to not use 'spin axis', but rather break it down into the component back spin and side spin.
    First, because the back spin component is what affect the vertical trajectory/path, whereas the side spin affect the draw/fade path. You can have the same amount of side spin, causing the same amount of draw/fade curve, and yet you will have a different spin axis because the back spin is greater/less. So their importance is independent, and I think spin axis muddies the water because it uses relative velocity. Looking at it yet another way, the same spin axis can result in very different ball flights because of the magnitude of the spin.

    Second yes, it has a single axis at any given time, however, can't that axis change orientation? The ballistics of ball flight is a little out of my range, but as the ball rotational velocity decreases, does it do so along that spin axis? IN other words, does the air equally affect back spin (almost always a much greater rotational velocity) as it does side spin? I'd guess that we see a greater decrease in 'back spin', and therefore the spin axis would change over time. Again, this is speculation.

  3. #33
    Join Date
    Jan 2012
    Posts
    4,096
    Quote Originally Posted by Mike Duffey View Post
    Jeff, the title of that thread was, "Johnny Miller in his prime, "second fire" in the downswing. Is this a movement you would teach? How would you guess this shows up in a kinematic sequence graph?".
    If Kelvin never meant for the second fire to related to pelvis angular velocity - which I'm sure he knows is what the kinematic sequence shows - then why did he title it that way? If he knew it was something different, then there is no purpose in asking if/how it would show up in the KS.
    Let's step back, because you and your colleagues seem to be missing the forest for the trees.

    Kel knew these movements did not show up in the kinematic sequence graphs produced by TPI. The pelvis lift is in the standard pelvis graph and the hip, knee and ankle extension AREN'T IN ANY GRAPH in the standard TPI report (as Phil Cheetham subsequently confirmed). Of course, Kvest has no sensors on the hips, knees and ankles and can't report this data.

    Of course, it became clear in our conversations your system does not report important movements of the hips involved in the "second fire", specifically internal and external rotation.

    Do you start to see Kel's point now? If the folks designing the systems and their reports don't know what to look for, or see important things that they don't think are important, their work will be materially flawed.

    I'm sure you are familiar with his Skype session with Tapio. You only need to work through the first five minutes to hear them going through the kinematic sequence. Tapio brings up the apparent second acceleration of the shoulders - which is very clearly and specifically a rotational velocity - and Kelvin refers to that as a "Second Fire". He makes no effort to clarify that he is talking about a non-rotational motion. A minute or so earlier, when the topic is pelvis rotation in the KS, he makes no effort to mention that his second fire would be a different kind of motion.

    In fact, in that thread when Kelvin was asked why he added some lines to the video, he responded,
    "What are the yellow lines for? To help those that cannot see 2D movement with their own eyes. It's up to the biomechanists to explain this movement to us. We just make observations. Where in the current graphs is this motion buried? Why do our observations not jibe with the deceleration shown in the graphs?"

    I have to think that Kelvin understands the different between rotation and linear translation. If not, he is an idiot who has no business teaching anything in golf. I believe he is not an idiot. So he keeps asserting that this motion should be seen in a graph of rotational velocity.
    He does not "keep asserting that this motion should be seen in a graph of rotational velocity". He asked: "Where in the current graphs is this motion buried?" Phil Cheetham admitted that these important movements ARE NOWHERE TO BE FOUND IN THE STANDARD TPI REPORT. So anyone teaching from those reports will be missing important information. And, as I read through that thread, none of you seemed to care.

    So, maybe he knew and maybe he didn't. Reading through the thread, it become eminently clear that the people in the conversation originally understood the Kelvin's second fire to be a rotational motion, not a linear one, and that understanding changed over time. Kelvin made no effort early on to clear up the topic.

    What is absolutely clear is that many of the early comments were made under the belief that Kelvin was discussing a second rotational velocity peak. My comment about not taking them out of context is that the discussion here, as it did in the original thread, has shifted to a linear motion. I agree that you have not reposted them, I'm trying to save anyone who starts reading this discussion now the headache that we went through as the conversation shifted from a rotational to a linear motion discussion.
    Thank you.

    I acknowledge that the current systems, other than 4dswing, don't show a second rotational peak, but am not convinced those systems are designed properly; I just don't know enough about them.

    As for putting together a good compilation of reading material, why not do it? It seems it would only help people understand the topic.

    I have for Dr. Kwon and the result was a huge disappointment. My feeling now is, why bother?

    To put it bluntly, you guys (golf biomechanists) are the equivalent of the person holding the stopwatch and clipboard at the race track, recording lap times. Kelvin and Lucas are the engineers who understand the design of the race cars and how each element of the design influences performance. Reams of information have been published by the three of us that explains all you need to know. Figure it out for yourselves.

    Did you see Lucas's recent case study of his student Ryland? Ryland's swing has been transformed in the course of a year by incorporating all the important movements Kel has identified. Ryland has taken his clubhead speed from the low 100s to 114-115mph (on Trackman), above the tour average, with a much more stable release than he had before, so any "look" of handle dragging hasn't hurt his speed. Of course, Lucas made a similar transformation with his own swing.

    My hypothesis is this: what Kel and Lucas teach works better than what anyone else teaches. To falsify my hypothesis, bring me a case study with results comparable to Ryland and Lucas. If you and your colleagues don't think it is worth your time to understand Kelvin's research, that's fine with me. As I told Dr. Kwon: good luck in your journey to identify the movements of an elite golf swing, it will be a long one.




    Jeff

  4. #34
    Join Date
    Jan 2012
    Posts
    4,096
    Art-

    I think you are making constructive contributions. Here is something I found in the TPI data base (but not in the reports, naturally) that I think is interesting, the speed of the pelvis throughout the swing, in mph. For those following at home, see this explanation of the difference between speed and velocity:

    http://www.physicsclassroom.com/clas...d-and-Velocity


    Name:  speed.PNG
Views: 604
Size:  281.8 KB





    hmmmm... What do you know? Two peaks in speed for the players with a meaningful second fire, with an earlier second peak for Els (Pro2) who doesn't lateral bend early. Paints a different picture than the rotational velocity graphs, which suggest that the movement of the pelvis slows during the second half of the downswing, even for a high clubhead speed player like Mr. Elite, when, in fact, the pelvis speeds up again.




    Jeff

  5. #35
    Join Date
    Jan 2012
    Posts
    4,096

    Some nice high speed video of the "second fire" in long drivers from REMAX...

    Of course, TPI tells us next to nothing about their "elite" movements: lift the left heel, vertical left arm, moving the feet, stuff a blind person can see...








    Jeff

  6. #36
    Join Date
    Jan 2012
    Posts
    4,096

    This is interesting...

    I grabbed this screen cap from the TPI/REMAX video. It appears to be the kinetic sequence of long driver Ryan Winther. Looks like his pelvis and thorax rotational velocities peak at the same time.

    Name:  ryan 1.PNG
Views: 546
Size:  407.4 KB





    Jeff

  7. #37
    Join Date
    Jan 2012
    Posts
    4,096
    Quote Originally Posted by Mike Duffey View Post
    In fact, in that thread when Kelvin was asked why he added some lines to the video, he responded,

    "What are the yellow lines for? To help those that cannot see 2D movement with their own eyes. It's up to the biomechanists to explain this movement to us. We just make observations. Where in the current graphs is this motion buried? Why do our observations not jibe with the deceleration shown in the graphs?"
    Let's take a look again at the final frame of the Johnny Miller gif:

    Name:  miller.PNG
Views: 983
Size:  936.5 KB


    Obviously, the movement of Miller's belt loop (marked by the yellow lines) is a combination of rotation (away from the target line) and translation (vertical lift). The circled spacing of the yellow lines from the last two frames does not suggest angular deceleration; in fact, it suggests angular acceleration. So, why is it unreasonable to expect this observed angular acceleration to be measurable? Especially when there exists a system (4dswing) which has reported this pattern?

    We have "found" the vertical lift movements thanks to Phil Cheetham, but what about this rotational component? Could it be your systems aren't designed to catch it, because you aren't looking for it hard enough?






    Jeff

  8. #38
    Join Date
    Aug 2012
    Location
    Pacific Palisades, California
    Posts
    73
    Quote Originally Posted by Jeff Martin View Post
    Art-

    I think you are making constructive contributions. Here is something I found in the TPI data base (but not in the reports, naturally) that I think is interesting, the speed of the pelvis throughout the swing, in mph. For those following at home, see this explanation of the difference between speed and velocity:

    http://www.physicsclassroom.com/clas...d-and-Velocity


    Name:  speed.PNG
Views: 604
Size:  281.8 KB





    hmmmm... What do you know? Two peaks in speed for the players with a meaningful second fire, with an earlier second peak for Els (Pro2) who doesn't lateral bend early. Paints a different picture than the rotational velocity graphs, which suggest that the movement of the pelvis slows during the second half of the downswing, even for a high clubhead speed player like Mr. Elite, when, in fact, the pelvis speeds up again.




    Jeff

    Jeff,

    Thanks for the kind comment; my ONLY purpose is to seek and find 'golf truth' to determine WHAT is right, and not interested in WHO, as there is no agreement YET. With regard to the exchange taking place in this thread, I am appreciative of everyone's efforts, and personally am learning a great deal from each post.

    Looking at this specific area of pelvic speed, I have to admit an uneasiness as to how the data are gathered. I believe the movement of the 'center' of the two greater trochanter sensors (or their electronic artifacts) moving thru the electromagnetic field may be what is being displayed. If so, without a clue as to the direction as it moves from one data point to the next data point, all it presently portrays to me is that the center of the pelvis is not still. But, at such low speeds, 0-2 miles per hour, IMO, it does not realistically represent the significant energy release shortly before impact. The work I have been involved in suggests looking at the lead hips trajectory (as I previously posted), and then ADD the contribution of the upward movement of the lead arm glenoid/humerus motion from the shoulder girdle to more realistically represent the location, direction and magnitude of the total 'second fire' event as both affect the trajectory of the grip and therefore CHS.

  9. #39
    Join Date
    Jan 2012
    Posts
    4,096
    Quote Originally Posted by art View Post
    Jeff,

    Thanks for the kind comment; my ONLY purpose is to seek and find 'golf truth' to determine WHAT is right, and not interested in WHO, as there is no agreement YET. With regard to the exchange taking place in this thread, I am appreciative of everyone's efforts, and personally am learning a great deal from each post.

    Looking at this specific area of pelvic speed, I have to admit an uneasiness as to how the data are gathered. I believe the movement of the 'center' of the two greater trochanter sensors (or their electronic artifacts) moving thru the electromagnetic field may be what is being displayed. If so, without a clue as to the direction as it moves from one data point to the next data point, all it presently portrays to me is that the center of the pelvis is not still. But, at such low speeds, 0-2 miles per hour, IMO, it does not realistically represent the significant energy release shortly before impact. The work I have been involved in suggests looking at the lead hips trajectory (as I previously posted), and then ADD the contribution of the upward movement of the lead arm glenoid/humerus motion from the shoulder girdle to more realistically represent the location, direction and magnitude of the total 'second fire' event as both affect the trajectory of the grip and therefore CHS.
    Agree 100%. Just thought it was interesting and certainly shows that the pelvis just doesn't go to sleep after peak rotational velocity.



    Jeff

  10. #40
    Mike Duffey Guest
    Quote Originally Posted by Jeff Martin View Post
    Let's take a look again at the final frame of the Johnny Miller gif:

    Name:  miller.PNG
Views: 983
Size:  936.5 KB


    Obviously, the movement of Miller's belt loop (marked by the yellow lines) is a combination of rotation (away from the target line) and translation (vertical lift). The circled spacing of the yellow lines from the last two frames does not suggest angular deceleration; in fact, it suggests angular acceleration. So, why is it unreasonable to expect this observed angular acceleration to be measurable? Especially when there exists a system (4dswing) which has reported this pattern?

    We have "found" the vertical lift movements thanks to Phil Cheetham, but what about this rotational component? Could it be your systems aren't designed to catch it, because you aren't looking for it hard enough?


    Jeff
    Jeff, it is technically possible that the 4D system 'does things better' than the existing systems; but it is highly unlikely. Optical systems have decades of use, accuracy testing, and validation. 4D does not. It does not mean that it is wrong, but some of the best minds in engineering and biomechanics have been trying to make that technology work, and so far, they have been unable. Because of Tapio's reluctance (that I know of) to publishing validation data, we won't know. But we can leave it as an option for the future.

    What might more likely be happening is that the 4D system is more susceptible/sensitive to soft tissue motion. Since they are doing shape recognition, and the 'shape' of our segments changes as we change joint angles, it is possible that the movement of of fat and muscle is being read as 'segment motion'. I believe this is going to be especially true as the trend toward more muscular/athletic golfers continues. If this is the motion the user is interested in, then this would not be considered an error - it is simply a different method of evaluating motion.

    However, there are some better things to discuss here. You asked why the 'rotational component' might be missed. I would flip that a bit and say there are some reasons why something might look like substantial rotation on 2D film isn't really a big rotation.

    1. Put a pencil on a table and hold the eraser win place with your right hand. Use your left hand and rotate the pencil about 20 degrees (for an 8 inch pencil, this means moving the end about 3 inches). If you have a camera off to the left, a lot of motion will be seen at the sharp end of the pencil. Now hold the sharp end still with your left hand and rotate the eraser end about 20 degrees. Now the camera to your left will show very little motion, even though the pencil rotated 20 degrees both times.
    2. You are going to start doing the same as above. You are going to move the pencil tip 3 inches with your left hand, but this time, instead of keeping the eraser still, move the eraser one inch in the same direction. The camera to the left still seems the same amount of motion, but the actual rotation drops from about 20 degrees to about 14 degrees. You could also do the above but move the eraser the opposite direction. Now you still have the same motion of the tip, but now you have about 30 degrees of rotation. What is happening on the other side is tremendously important in this, and most of that motion is hidden because it is virtually impossible to know what the right hip is doing during the swing because you would have to see through Johnny's body.

    Your question was an excellent one, it is something that many in the golf instruction world who have a lot of experience with 2D work may need help understanding.

  11. #41
    Mike Duffey Guest
    I agree with Art's comment about the magnitude of the speed being so low that there is not much energy to be gained. Even more so because there is a lot of distance between the pelvis and the club and very little time to send that energy up the chain to make it useful.

    What may be more important than the magnitude of the speed itself, is the rate of change. Any acceleration will be a direct and proportional result of force; so the accelerations and decelerations will indicate a change in the applied force. How and when these are applied can make a very big difference in how the club is delivered going into impact. You have mentioned before that you think very little of instructors who try to teach using force, but this may be a case where an instructor is trying to teach a movement (a small change in pelvis velocity), but what is really important is that motion creates the necessary conditions for the all-important force at the right time.

  12. #42
    Mike Duffey Guest
    Quote Originally Posted by Jeff Martin View Post
    hmmmm... What do you know? Two peaks in speed for the players with a meaningful second fire, with an earlier second peak for Els (Pro2) who doesn't lateral bend early. Paints a different picture than the rotational velocity graphs, which suggest that the movement of the pelvis slows during the second half of the downswing, even for a high clubhead speed player like Mr. Elite, when, in fact, the pelvis speeds up again.

    Jeff
    Jeff, the rotational velocity graphs show that rotational velocity slows. That is the kinematic sequence. The graph of linear velocity very well might show something different. No one that I know of is strongly claiming that the linear velocity has to be slowing - partly because the linear velocity can be very, very different for different parts of the same segment. That is exactly the pencil example above.

    I don't know why you would even mix the two velocities in the same sentence. They can and often are, entirely different. Phil agrees with you that linear velocity might increase again, and I think he can speak for TPI, so TPI agrees with you. I agree with you.

  13. #43
    Lukman Ahmed Guest
    Great stuff, art. So nice to hear from you again.

    And Jeff--thank you for continuing to quantify for the biomechanists in their own realm the qualitative analysis you, Kelvin, Rick, and Luke have done.

    The collaborative effort in this thread is awesome.

  14. #44
    Join Date
    Jan 2012
    Posts
    4,096
    Kel posted this for me on the Golf Biomechanists Facebook page before he got the boot, seeking comment from Sasho Mackenzie, who published the paper where it came from. Sasho never responded. Perhaps Professor Duffey or Art would like to take a crack at it.

    In 2009, Mackenzie published this paper:

    http://people.stfx.ca/smackenz/Publi...golf swing.pdf


    The underlying study used a model of a golfer with four torque generators:

    1) the rotating torso (presumably powered by the pelvis, core muscles and shoulders)

    2) the shoulder joint (powered by lead arm transverse abduction and the straightening of the trail arm)

    3) the arm (powered by lead shoulder external rotation)

    4) the wrist (powered by ulnar deviation)


    The study sought to determine the optimal pattern of applying the four torques to maximize clubhead speed:

    Name:  Screen Shot 2014-05-16 at 9.29.08 PM.png
Views: 309
Size:  21.3 KB



    This figure summarizes the optimization:

    Name:  torque.PNG
Views: 397
Size:  207.7 KB


    As you can see if you look closely at the bottom of the three graphs, the torque applied by the torso peaks at impact.

    We know that torque = (moment of inertia) x (angular acceleration); we also know that angular acceleration = (torque)/(moment of inertia). So, for the torque generated by the torso to be increasing at impact, the angular acceleration of the torso must also be increasing (assuming the moment of inertia of the torso doesn't decrease during the downswing, which we can safely assume).

    So, Mackenzie's optimized model finds that, to maximize clubhead speed, the torso should be accelerating at impact. But, that blows up the kinematic sequence theory, doesn't it? And contradicts the kinematic sequence graphs we see (other than from 4dswing) which all show deceleration of the torso pre-impact.

    Can someone explain what's going on?


    Jeff

  15. #45
    nmgolfer Guest
    Jeffy,

    First... you got it wrong... Torque = Moment of Inertia X Angular acceleration

    Second ... the obvious conclusion is that the kinetic chain theory is crap (as you suggest)

    MacKinzies chart clearly shows a "second fire" of the torso torque.

    Carry on

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •